Quick Start

Python
# Generate 100 realistic user records
import faker
fake = faker.Faker()

users = [
    {
        "id": i,
        "name": fake.name(),
        "email": fake.email(),
        "created_at": fake.date_time_between(start_date='-2y'),
        "country": fake.country_code()
    }
    for i in range(1, 101)
]

Workflow

1. Analyze data requirements - Identify fields, types, constraints, relationships
2. Choose generation strategy - Faker library, custom generators, or pattern-based
3. Define realistic constraints - Date ranges, value distributions, foreign keys
4. Generate and validate - Create data, check for realistic patterns
5. Export in target format - JSON, CSV, SQL inserts, or API fixtures

Progress:

• Map required data schema
• Set up Faker providers or custom generators
• Define realistic business rules and constraints
• Generate sample dataset
• Validate data quality and export

Examples

Example 1: Input: "Need 50 e-commerce orders with products, customers, and timestamps from last 6 months" Output:

Python
orders = [
    {
        "order_id": f"ORD-{fake.random_number(digits=6)}",
        "customer_email": fake.email(),
        "items": [
            {"product": fake.catch_phrase(), "price": fake.random_int(10, 200)}
            for _ in range(fake.random_int(1, 4))
        ],
        "total": sum(item["price"] for item in items),
        "created_at": fake.date_time_between(start_date='-6m')
    }
    for _ in range(50)
]

Example 2: Input: "Mock API responses for user authentication flow" Output:

JSON
{
    "login_success": {"token": "eyJ0eXAi...", "expires": "2024-01-15T10:30:00Z"},
    "login_error": {"error": "invalid_credentials", "message": "Email or password incorrect"},
    "profile": {"id": 1847, "name": "Sarah Chen", "role": "admin"}
}

Best Practices

• Use Faker library for consistent, locale-aware data generation
• Maintain referential integrity (consistent user IDs across related tables)
• Include edge cases: null values, empty strings, boundary conditions
• Generate data in realistic distributions (80/20 rules, seasonal patterns)
• Seed random generators for reproducible test datasets

Common Pitfalls

• Creating unrealistic data patterns (all users born on same day)
• Ignoring business constraints (negative prices, future birth dates)
• Generating too much data for development environments
• Using production data patterns that reveal real information